There's plenty of fear and loathing out there about "rogue" services that can bring enterprises down (or at least slow things down), and governance tools vendors are starting to make a pretty good living promising to keep such beasts at bay. Users shouldn't be allowed to run amok and build their own services that could clash with the carefully planned architectures that run mission critical processes, right?
Well, David Margulius just posted this interesting account of a doctor that was unhappy with his medical center's IT system, which had an electronic inpatient medical records system, but not one for outpatient records. So, he took it upon himself to jury-rig his own outpatient medical records system.
I'm sure the hospital's HIPAA managers and lawyers would go ballistic if they learned that a doctor took it upon himself to build his own system to manage medical records. But, at the same time, David noted that a competing hospital chain was spending billions of dollars on an "official" outpatient records system that probably doesn't do much more than the one the doctor pieced together in his spare time.
In another account I just saw, a CIO at the State University of New York credits "rogue" services with reshaping the school's distance learning system as well as other online services. "Look at the wiki and the blog. Look at Moodle’s success – how many of those deployments started life as an official top down implementation versus how many started life as a rogue implementation by an individual academic in their department?" asked Patrick Masson, CIO at SUNY's Delhi School of Information Technology. "That’s the environment we’re living in and I think an agile development approach and a SOA approach fits that real world academic computing environment."
Perhaps there's something to be said for giving end-users should be given some latitude to experiment and come up with their own solutions. Enterprise architecture and a governance structure are vital to maintaining a viable and workable IT infrastructure, but with Enterprise 2.0 and Web 2.0 approaches swirling about, there will always be groups doing their own workaround solutions to problems. Readers, what do you think?